Historically, most transaction systems use relational database platforms with a normalized data model to efficiently store data and manage updates. While the relationships stored in the normalized data model are the basis for hierarchical consolidations, transaction systems tend to provide form-based or tabular views to users for viewing and updating of information. This architecture is highly efficient for transaction processing and integrity.
Normalization is the process of structuring relational database schema to remove most redundancies and inconsistencies. The stages of normalization are referred to as normal forms, ranging from the least restrictive first normal form to the most restrictive fifth normal form. Generally, each higher normal form must meet all the requirements of each lower normal form in addition to meeting the specific requirements of the higher normal form. Most relational database designs meet the requirements of the third normal form (3NF) rules in which all columns are removed that do not depend on the primary key. In other words, all non-key attributes are mutually independent.
Many computer software applications for reporting and analysis, such as data warehouses and Online Analytical Processing (OLAP) applications, use hierarchical representations of data relationships (hierarchies) to support user visualization of the relationships and provide a navigation path from summary data to detail data and vice versa. Current software applications that provide both hierarchical visualization and hierarchy editing capabilities to users rely on specialized or proprietary data stores which often require duplication of data from transactional or other systems.